Conformable Vision-Based Tactile Sensor with Enhanced Soft Elastomer Design for Palpating Irregular Anatomical Surfaces

Many cartilage tissues in human organs are covered by skin, making it difficult to assess their characteristics through visual observation alone. Tactile sensing can provide complementary information in such scenarios. Vision-based tactile sensors (VBTS) offer ultra-high resolution and are capable of perceiving relatively large contact areas. In this work, we investigate the application of VBTS in surgical robotic palpation tasks. We enhance the existing sensor hardware to improve its compatibility with the uneven and skin-covered surfaces of cartilage tissues. Finite element simulations are conducted to analyze the underlying mechanical interactions of palpation. Furthermore, we integrate the improved sensor with a robotic arm and perform palpation experiments. The results demonstrate that the VBTS effectively captures both shape and stiffness information of the throat region. Compared to the original design, the enhanced sensor module achieves superior tactile perception performance.